SIMON'S CROMWELL CROWN DIES IN THE ROYAL MINT MUSEUM AND BLONDEAU'S METHOD FOR THE PRODUCTION OF LETTERED EDGES

Transcription

1 SIMON'S CROMWELL CROWN DIES IN THE ROYAL MINT MUSEUM AND BLONDEAU'S METHOD FOR THE PRODUCTION OF LETTERED EDGES PETER P. GASPAR FOR the student of English coinage technique there is no more interesting entry in Hocking's catalogue of the die collection in the Royal Mint Museum than the items number 18 and 19, whose descriptions follow: Crown. Matrix. Obv. (by Simon), laureated and draped bust of Cromwell to left, different from punch No. 16 and from Simon's coins; OLIVAR.D.G.RP.ANG.SCO.HIB &C. PRO. 19. Crown. Matrix. Rev., 1658 (by Simon). Crowned and garnished shield of arms, differing in details from known coins: PAX QVAERITVR BELLO, The significant position of these instruments in the history of English die-making is that if they were correctly identified they would be the earliest known matrices bearing complete designs, including lettering and date. Hocking lists twenty further lettered matrices for the following century, 2 but the earliest lettered punch to survive is from the last quarter of the eighteenth century or the second decade of the nineteenth century. 3 If Hocking's identification of the Simon Cromwell crown matrices and of the twenty other lettered matrices of the following century is accepted, then we must conclude that the intention existed to eliminate the practice of lettering dies with individual letter punches long before it was finally practical to do so. The absence of lettered punches for the period in the museum of the Royal Mint, while more than 200 unlettered punches survive from this period, would suggest that it was not feasible to raise punches from the twenty-two lettered matrices, or that perhaps by sheer chance no lettered punches have survived. The latter view is certainly unlikely, considering the large number of heavily used unlettered punches which remain. If the production of lettered punches had been within the realm of technical feasibility, the great saving in time which their use would have effected makes it unlikely that so many unlettered punches would have been prepared and put into use. It is the thesis of this paper that, with all due respect to Hocking, the objects identified as Simon's Cromwell crown matrices are actually dies, the very dies that struck the original Cromwell crown pieces attributed to Simon. 4 This proposition is not a new one, 1 W. J. Hocking, Catalogue of the Coins, Tokens, Medais, Dies and Seals in the Museum of the Royal Mint, vol. ii, Dies, Medals and Seals, HMSO, London, 1910, p Ibid., nos. 94, 99, 237, 324, 326, 353, 356, 360, 366, 373, 390, 393-6, 399, 401, 404, 406, and Ibid., no. 421 is a lettered punch for the reverse of a pattern guinea by L. Pingo. No. 423 is a lettered obverse punch for the Military Guinea, also by L. Pingo. 4 For a recent and thorough summary of the Cromwell coinage see: M. Lessen, BNJ xxxv (1966), 163. These crown dies were among the nineteen punches and dies purchased by Isaac Newton, Master of the Mint, in The purchase is reported in the minutes of the 9 Nov meeting of the Board of the Royal Mint and in an item dated 25 Dec in the manuscript record of Statements of the Warden's and Master's Accounts kept by Newton. See: S. A. H. Whetmore, BNJxxx (1960), 159.

2 56 SIMON'S CROMWELL CROWN DIES IN THE ROYAL MINT MUSEUM AND but was endorsed by Vertue, 1 Ruding, 2 and Henfrey. 3 Its resurrection requires us to re-examine the method by which the edge inscription was applied to the Cromwell crown. It will be argued that for both the Cromwell and Petition crowns, as for all English coinages with which Blondeau may have been associated, 4 the edge inscription was applied in a separate operation prior to the striking of the obverse and reverse, and that the final striking was carried out without the benefit of a collar. This idea too is not a new one, having been suggested by H. G. Stride in 1957, 5 but we shall see that previously held views about the manner in which the edge inscription was applied to the Cromwell crown require modification. A comparison of photographs (PI. IX, 1 and 2) of impressions from the alleged crown matrices (obverse shown in PI. IX, 3) with photographs of the Simon crown pieces (PI. IX, 4 and 5) offers vivid support to the thesis of this paper. The obverse 'matrix' carries the identical pronounced die break present in some stage of its development on all surviving coins of the issue attributed to Simon. 6 In all respects the obverse and reverse 'matrices' appear identical with the Simon coins. 7 Marvin Lessen has written of the impressions from these 'matrices' illustrated by Hocking: 8 'I am unable to discern any differences between this impression and a normal Simon striking, at least nothing which could not be explained by the condition of the steel.' 9 It is tempting to dismiss the description in Hocking's great catalogue of what we must now call the Simon Cromwell crown dies as an error caused by the shape of these objects they resemble modern matrices in their lack of a neck. However, Hocking has written a comprehensive account of the Cromwell dies, and we must therefore examine Hocking's arguments with care. Two principal objections were raised by Hocking to the earlier identification of the objects with which we are concerned as the dies which struck the Simon Cromwell crowns: 10 (1) 'They have no neck like dies, and could not easily be used to strike coins.' (2) 'The inscriptions are also different in style from those on Simon's coins, none of the letters having identations at the bases.' As to the first of these points, the slight curvature of the surface outside of the rim beading does certainly preclude the use of these objects as dies together with a collar, and our discussion of the second point will establish beyond doubt that the Simon Cromwell crowns were struck without simultaneous use of a collar. Hocking states that the bases of both 'matrices are out of parallelism with the faces' and that 'this would make it almost impossible to obtain a good impression on them from the master 1 G. Vertue, Medals, Coins, Great Seals, and Other reverse 'matrices' with photographs of a specimen Works of Thomas Simon, 2nd edn., J. Nichols, of the Simon Cromwell crown (in the British Museum) Londpn, 1780, p. 78. that shows the die break distinctly. This photographic 2 R. Ruding, Annals of the Coinage of Great overlaying comparison technique facilitates very Britain and its Dependencies, 3rd edn., John Hearne, accurate comparisons of design elements on different London, 1840, vol. II, p. 336, note. numismatic specimens and allows detection of posi- 3 H. W. Henfrey, Numismata Cromwelliana, John tional variations greater than c mm. Its applica- Russell Smith, London 1877, p tion leaves no doubt concerning the identity of the 4 But see A. J. Nathanson, Thomas Simon, his Life die crack carried by the Simon Cromwell crowns and and Work, , London, the crack on the objects Hocking called matrices. 6 H. G. Stride, 'The Royal Mint', part XI, Seaby's The photographic overlaying technique is described Coin and Medal Bulletin, July 1957, p by its inventor: J. A. Haxby, Transactions of the 6 H. W. A. Linecar and A. G. Stone, English Proof Canadian Numismatic Research Society, vii. 42 and and Pattern Crown Size Pieces, Spink & Son, London, 100 (1971). 1968, p W. J. Hocking, Numismatic Chronicle, 4th ser., 7 Minute comparison was carried out by Dr. ix (1908), p Lessen, op. cit., p James Haxby of photographs of both obverse and 10 Hocking, 1908, pp

3 BLONDEAU'S METHOD FOR THE PRODUCTION OF LETTERED EDGES 57 puncheons or to transmit impressions from them to working punches.' 1 Hocking does, however, attribute a coin in the Hunterian Museum to the use of the obverse 'matrix' as a die muled with Tanner's reverse die. 2 Turning to the second point it is certainly true, as Figures 1 to 5 reveal, that the bases of most letters both on the obverse and reverse do have a different appearance on the Simon crown pieces and on the objects we identify as the dies which struck the coins. Rubber impressions from the dies resemble the dies themselves rather than the coins in the details of the lettering. Hocking noted that on the objects he identified as matrices and on the impressions he had taken from them, there are no indentations at the bases of the letters, while the coins show these indentations or bifurcations clearly. This apparent discrepancy is explained by the formation of indented or bifurcated letter bases on the coins due to a radial outward flow of metal during the striking of a blank in a mechanical press without the use of a collar to restrain the outward metal flow. C. W. Peck has pointed out that the curved serifs and bifurcated letters that were believed by Hocking to be a stylistic feature of Simon's work, 3 were in fact a consequence of striking coins under the pressure provided by mechanical presses without constraining the blank within a collar. 4 These bifurcations persisted until the use of collars was introduced and are still evident in such ungirdled coinages as the Maria Theresa talers. 5 A comparison of coins struck from similar dies with and without the use of a collar provides a vivid demonstration of this effect. PI. IX, 6 shows part of the obverse legend on an 1816 shilling (ESC 1228). PI. IX, 7 shows the same portion of the obverse legend on a shilling of the same date struck from a similar die on a flan that by accident was not enclosed within a collar. PI. IX, 8 shows the entire obverse of this latter piece. All the letters on the piece struck without a collar show the indented bases which are totally absent on the normally struck piece. The absence of bifurcated letter bases from the modern impressions taken from the Simon crown dies is due to the nature of the substrate 6 and striking pressure, the modern impressions having been taken gently and gradually in soft materials and therefore perfectly mirroring the dies, showing no distortion of the lettering and hence straight letter bases. Hocking also employs the presence of bifurcations on the so-called fifty-shilling piece of Cromwell as evidence that they were struck from different dies from those that struck the Cromwell broads on which bifurcations are barely discernible. 7 Lessen concurs with Henfrey in stating that both coins were struck from the same dies. Lessen has suggested that the bifurcation appears to be a function of the striking process and is possibly related to the thickness of the metal used. 8 The thicker the blank, the greater 1 Ibid., p the misapprehension that bifurcated lettering on dies 2 This piece is no. 24 in Lessen's catalog of the is necessary for the appearance of bifurcated letters Cromwell coinage. on coins is afforded by the observation that modern 3 Hocking wrote of Simon's work (Hocking 1908, London dies for the Maria Theresa talers were prop. 97):'His letters are for the most part distinguished duced with slight bifurcations on some letters by a bifurcation or indentation at the bases of the always less pronounced, however, than those that vertical bars a form of ornamentation of frequent appear on the coins. occurrence since that date.' 6 Lessen, op. cit., p. 169, refers to the impressions 4 C. W. Peck, English Copper, Tin and Bronze Coins illustrated by Hocking as possibly being of wax. in the British Museum, , 2nd edn., London, The impressions illustrated in PI. IX, 1 and 2 are 1969, p See also K. Bressett, The Colonial silicone rubber direct castings. Newsletter, vol. 2, no. 4, Oct.-Dec Hocking, 1908, p An amusing indication of how widespread was 8 Lessen, op. cit., p. 166.

4 58 SIMON'S CROMWELL CROWN DIES IN THE ROYAL MINT MUSEUM AND the ease of metal flow that is to be expected, and the extent of bifurcation should increase with increasing outward flow of metal. 1 Only minor variations in the appearance of the letters have been noted by Lessen, however, when it was possible to compare strikings in different metals from the same die or in one metal but of different thicknesses, all in the series of Cromwell coins. 2 We are therefore forced to dismiss the distinction based on bifurcated letters between the objects identified by Hocking as matrices and the dies from which the Cromwell crowns were struck. We now know that Hocking's 'matrices' were the original Simon dies. The bifurcations seen on the coins are significant because they provide positive evidence that the Simon Cromwell crowns were struck without simultaneous use of a collar. The presence, on the object described by Hocking as the obverse matrix, of precisely the same die crack across the lower portion of the bust as is found on the coins, proves beyond doubt that it was indeed the die from which the original Cromwell crowns were struck. Careful comparison of the positions of all design elements including the inscriptions on both the obverse and reverse dies with those on the coins leads to the same conclusion. Since coins are known with varying states of the same characteristic obverse die crack (but it is believed that no coins exist without a trace of the crack) the only way in which the objects under consideration could really have been matrices is if a series of punches had been raised which sank a series of dies with varying states of the obverse die crack. This theoretical alternative 3 is precluded by the unlikelihood of successfully raising several punches from a cracked matrix. One punch could, moreover, produce enough dies for the striking of a much larger issue of coins than that of the Cromwell crown. The author is not aware of evidence aside from their listing by Hocking for the use of fully lettered matrices before the end of the eighteenth century. It is suggested that all twenty-two of the objects described by Hocking as being earlier lettered matrices are dies. This generalization is based upon careful examination of the remaining twenty instruments and comparison with surviving tools known to be dies, and with coins. Existing records do, however, indicate that matrices were indeed produced before These seem to have been impressions of device and portrait punches made by the engravers. Whether these early matrices were actually employed to raise punches is not definitely known. 5 If the twenty-two early coining implements listed by Hocking as matrices are regarded as dies, as this paper suggests, then the earliest surviving matrices may be sought among the objects listed by Hocking as unlettered matrices of the late seventeenth and eighteenth centuries. 6 The earliest object in the die collection of the 1 F. R. Barton, Nickel Coinage, Mond Nickel Company, n.d. (1926), p Private communication from M. Lessen, Suggested to the author by Dr. Alfred P. Wolf. 4 For a warrant from Charles II to the Roettiers for the engraving of 'Puncheons, Counter Puncheons, Mattrices and Dyes' and for the receipt given by the Roettiers for payment, see: J. K. R. Murray, BNJ xxxviii (1969), p Pertinent to this question is the author's current investigation to determine whether the farthing reverse dies for the reigns of George I and II and for the beginning of the reign of George III were sunk from a single punch as suggested by Peck (op. cit., pp. 142, 201, 207, 212, 232) or were instead the products of several punches raised from one or more matrices. 6 Hocking, 1910, nos. 255, 306, 328, 329, 386, 387, 411, 412, 415, 417, 510, 512, 514. Some of these tools appear upon examination to be unfinished dies, e.g. nos. 255, 328, 329, 386, 387, 512, 514. While 306 may be a matrix, it belongs to the reign of George III, not Anne. No. 510 does appear to be a matrix for an Irish halfpenny obverse of 1781.

5 BLONDEAU'S METHOD FOR THE PRODUCTION OF LETTERED EDGES 59 Royal Mint Museum presently recognizable with complete certainty as a matrix is a lettered guinea obverse matrix of If the thesis is accepted that the subjects of this paper are the original Simon Cromwell crown dies, then the edges of the crown pieces were not marked with the inscription *HAS-NISI-PERITVRVS-MIHI-ADIMAT-NEMO by the blank being held in a collar while struck with the obverse and reverse dies. This is a necessary conclusion since the objects we have identified as dies for the Cromwell crown could not, because of their shape, be used with a collar. We have seen that this conclusion is reinforced by the occurrence of bifurcated letter bases on the coins. Unfortunately this conclusion once again draws us into conflict with a deduction made by Hocking. Hocking has described the competition between the Company of Moneyers and Peter Blondeau held in 1651 in which patterns were prepared to determine the manner in which lettering and graining would be applied to the edges of coins to be produced by the screw press. 2 It was suggested by Hocking that David Ramage, the champion of the Moneyers, marked the edges of his coins by means of a segmented collar. Hocking quotes Blondeau as stating that this method destroyed many dies and collars because the screw of the press was not sufficiently accurate to prevent the upper die from occasionally striking the collar upon its descent. Hocking believed that Blondeau, to lessen the consequences of this problem, substituted a thin strip of engraved steel for the segmented collar. The use of a stencil-like steel strip held in a collar to apply edge lettering was believed by Hocking to constitute the 'new invention, not yet practised in any state in the world', that Blondeau himself claimed. 3 The thin strip of steel, according to Hocking, carried the inscription with which the edge of the com was to be marked as a series of perforations shaped as letters. The steel strip (or several overlapping strips) was placed within a smooth collar, and the blank to be coined was placed within the steel strip. When the blank was struck with the dies it would expand through the perforations of the inscribed band and be contained by the smooth collar. The edge would thus be marked with raised lettering whose relief corresponds to the thickness of the inscribed band. The coin and band together would finally be knocked out of the smooth collar, and the band would spring free leaving the finished coin. The virtue of this method, according to Hocking, was that the bands were easier and cheaper to make than the segmented collars, and therefore their loss upon being mistakenly struck and shattered by the upper die was less serious than the loss of a segmented collar. The Cromwell crowns, half-crowns, 'fifty shilling piece', and broad were believed by Hocking to have had their edges marked by this method at the same time that the obverse and reverse impressions were applied. Simon's later Petition Crown was believed by Hocking to have been produced in the same manner. 1 Hocking, 1910 p. 27, no Henfrey describes the competition in greater detail. See Henfrey, op. cit., pp For biased but contemporary views, see T. Violet, The Answer of the Corporation of Moniers in the Mint, at the Tower of London, to two false and scandalous Libells printed at London, and lately come forth without date, Printed for the Corporation of Moniers, Violet includes in his work 'A most humble memorandum from Peter Blondeau concerning the offers by him made to this Commonwealth for the coyning of the monie, by a new invention not yet practised in any state of the world...'. It has been suggested that Blondeau's coining press may have been a drop press rather than a screw press: C. L. Mason, Seaby's Coin and Medal Bulletin, May 1948, p In Blondeau's 'A most humble memorandum...' reprinted by Violet, op. cit.

6 60 SIMON'S CROMWELL CROWN DIES IN THE ROYAL MINT MUSEUM AND In his description of the method by which the edges of the larger Cromwell pieces and the Petition Crown of Charles II were lettered, Hocking is in complete accord with George F. Ansell, who described the process similarly in his book, The Royal Mint. 1 Ansell, who was employed in the coining department of the Royal Mint from 1856 to 1868, attributed the description of the steel band used to letter the edge of the Petition Crown to W. H. Barton, Deputy Master. Ansell noted that the edge of the specimen of the Petition Crown in the Royal Mint museum is slightly concave, and suggested that this was a deliberate means taken to protect the inscription, the famous petition. Hocking suggested that the segmented collar was replaced by the steel band in order to decrease the loss suffered when the descent of the upper die was not perfectly centred within the collar, and therefore die and collar collided, destructively. When the upper die struck and shattered the thin steel band, the latter could be easily and inexpensively replaced. The die would, however, also strike the massive outer collar which encompassed the steel band (and the blank within) and would be damaged as surely and as severely as by the segmented collar. Frequent replacement of the upper die would have been necessary, although this would perhaps have been less costly and time-consuming to replace than the segmented lettered collar. Therefore it is reasonable to suggest that a method was sought for lettering the edges which minimized the destruction of engraved dies as well as engraved collars. One method, although quite slow, would have been to use the inscribed thin steel bands as described by Ansell and Hocking together with blank dies as suggested by Stride. 2 Stride quotes the 1662 contract with Blondeau: 'He is to make all Gold and Silver pieces round before they are sized and to marke ye Edges of all the Gold and Silver Coynes with Letters or Graynings according to their respective Sizes before they are Stamped upon the flat sides. He is likewise to prepare the Rings or Viroles of Steel for marking the Edges of the Severall Coynes with Letters and Graynings.' 3 The striking of the Petition and Reddite Crowns without a collar is indicated by the observation that the neck of the surviving obverse die 4 has a larger diameter (1-639 to inches) than that of the coins struck from it. The specimen of the Petition Crown in the Royal Mint museum has a diameter found to vary between and inches, depending on the position of measurement. Its companion Reddite Crown has a measured diameter between and inches. 5 Since the diameter of the die is between 0-05 and 0-11 inches greater than that of the coins, the die, although provided with a neck, could not have fitted within a collar which simultaneously provided the edge lettering. It is interesting to note that the Petition and Reddite Crowns do not display bifurcated letter bases. However, while the presence of bifurcations strongly suggests that no 1 G. F. Anseli, The Royal Mint, 3rd edn., London, 1871, pp Stride, op. cit., 5 Ibid. p Stride's quotation (unreferenced) is from the 18 Apr order of the Privy Council Board on Mint Affairs, Mint Record Books, Mint 1/1, p The same order (p. 144) tells us that Blondeau is '... to prepare and make ready... the Engines to mark the Edges of the Moneys with Letters and Graynings, the great presses for Coining of Moneys...' If the inference is drawn that the edge-marking apparatus is less massive than the coining presses, then this speaks against Stride's suggestion. Application of edge lettering by striking with blank dies and using an inscribed band within a solid collar would require presses as powerful as those employed with engraved dies. 1 Hocking, 1910, 1, p. 11, no These measurements were kindly provided by G. P. Dyer, Librarian and Curator of the Royal Mint.

7 BLONDEAU'S METHOD FOR THE PRODUCTION OF LETTERED EDGES 61 collar was used during striking, the absence of bifurcations does not prove that a collar was employed during striking. The absence of bifurcations may be due to multiple impressions of the die on the blank, as in the production of medals or modern proof coins. A single striking at increased pressure could also produce a coin without indented letter bases. 1 Examination of the rims of Petition Crowns in the British Museum and the Royal Mint leads the author to believe that the edge lettering was applied before the blank was struck with the obverse and reverse dies, rather than afterwards. The edge lettering on the Petition Crown may have been applied as first suggested by Ansell, with the thin engraved steel band acting as a stencil through which the blank was forced to expand by the pressure exerted by blank dies. The single curved 'witness line' which follows the words 'more' and 'him' 2 was cited by Ansell as early as 1862 as being due to the use of this device. 3 Hocking proposed that the edges of the Cromwell < TjTr\ o r \\ crown, half-crown, 'fifty shilling piece', and broad were '» ^ K b J lettered with a similar device. Close examination of the \ H I * JJ coins suggests instead that the edge lettering was rolled - on to all the Cromwell coins of Simon by a precursor of the Castaing machine, which Hocking believed to have been used only for the Cromwell shilling. 4 Examination of the Blondeau pattern half-crowns of 1651 revealed that the same method was used! Hocking has argued that the vertical lines surrounding two dots in the edge inscription on the Simon Cromwell crown: HAS NISI PERITVRVS MIHI ADIMAT NEMO * represents the overlapping edges of the steel ribbon used to letter the edges and that the dots between the vertical lines indicate the positions of guide marks or catches which secured the two segments of the inscribed ribbon uniformly with respect to each other. 5 This arrangement may be represented as follows: outer view of overlapping ribbons with 'rear' ribbon 0') I 1 i\va r S indicated by dotted lines I JV X I I I 1 J> side view of overlapping ribbons Pl. IX, 9 and 10 are photomicrographs (at x6 magnification) of the two areas on the edge of a Simon Cromwell crown where vertical witness lines surround dots. The 1 The author is grateful to Messrs A. Crowhurst 2 For illustrations of the edge inscriptions of the and G. Johnson of the Royal Mint, Tower Hill, for Petition and Reddite Crowns, see Linecar and Stone, pointing out that high striking pressures could cause op. cit., p. 9. an apparent doubling of letters. Multiple striking is 3 G. F. Ansell, A Treatise on Coining, J. S. Virtue, of course another source of doubled lettering, and London, 1862, p. 25. This is the work to which Ansell these two quite different processes are not always later refers as the first edition of The Royal Mint. readily distinguishable from the appearance of the 4 Hocking, 1908, p. 56. coins. A Reddite Crown in the British Museum shows 6 Ibid., p. 91. doubling of the letters of CAROLVS on the obverse.

8 62 SIMON'S CROMWELL CROWN DIES IN THE ROYAL MINT MUSEUM AND important features to be noted are: (a) The dots are both doubled. (b) In Figure 10 there is a raised region where the dots overlap, (c) The left-hand line is stronger than the right in both PI. IX, 9 and 10. Features a and b are not compatible with the use of overlapping steel bands. If one of the dots of a doubled pair were on a rear band behind the band carrying the other dot, we would expect to see clearly defined on the coin only one dot and the region where the dots overlap. Instead in PI. IX, 10 we see both dots and the overlap area clearly defined. These features are easily explained if the edge inscription was applied t overlap area by rolling a blank between two parallel bars inscribed HAS NISI PERITVRVS and MIHI ADIMAT NEMO * respectively. The overlap may have occurred because both bars were longer than half the circumference of the blank or perhaps because the blanks underwent skidding while being passed between the bars. Such overlapping inscriptions are common on the milled crowns and half-crowns of the period PI. IX, 11 shows such an overlap region on the edge of a 1700 crown. The vertical marks closely resemble those of PI. IX, 9. Overlap is indicated by the bisection of the right dot in PI. IX, 11. It is therefore suggested that the edges of the Simon Cromwell crowns (and the half-crowns, 'fifty shilling pieces' and broads as well) were lettered with an edge-marking device similar to the later Castaing machine rather than with inscribed steel bands held within an outer collar as suggested by Hocking and later by Stride. It is clear that Blondeau's method for the production of lettered edges involved the use of a device like the later Castaing machine. We are not yet in a position to answer the question whether Blondeau's method was his own invention, or whether the extreme secrecy with which he guarded it was intended to conceal its source. In summary, the available evidence supports the view that the dies from which the Simon Cromwell crowns were struck are preserved in the Royal Mint museum, having been mistakenly catalogued as matrices in The Cromwell pieces of Simon seem all to have been struck without simultaneous use of a collar, and the edge lettering of the larger pieces was rolled on in an operation separate from, and prior to, striking with a device similar to the later Castaing machine. Acknowledgement. I am very grateful for the interest in this subject and the extensive advice and assistance of Mr. Graham Dyer of the Royal Mint. This subject could not have been developed without his patience and enthusiasm. I thank Dr. John Kent of the British Museum for encouragement and for gracious hospitality in the Department of Coins and Medals. For continuing advice and for the loan of a set of Cromwell pieces for study and photography I thank Mr. Eric P. Newman. For photographic assistance and for inspiration I thank Dr. James Haxby of the Bank of Canada. Discussions with Mr. Marvin Lessen, the unrivalled expert in the coins of Cromwell, were

9 BLONDEAU'S METHOD FOR THE PRODUCTION OF LETTERED EDGES 63 most helpful. Photographs of a Cromwell crown were kindly supplied by the American Numismatic Society, New York. Photographs of the rubber castings of the crown dies were kindly taken by Mrs. Phoebe D. Weil. ILLUSTRATIONS (PI. IX) 1. Impression from the obverse 'matrix' (Royal Mint) 2. Impression from the reverse 'matrix' (Royal Mint) 3. Obverse of the alleged matrix (Royal Mint) 4. Obverse of a Simon Cromwell crown (American Numismatic Society) 5. Reverse of a Simon Cromwell crown (American Numismatic Society) 6. Partial obverse legend, 1816 shilling, normally struck (Bank of Canada) 7. Partial obverse legend, 1816 shilling, struck without collar (Bank of Canada) 8. Obverse of 1816 shilling, struck without collar (Bank of Canada) 9. Partial edge inscription (magnification x 6) of Simon Cromwell crown (Eric P. Newman) 10. Partial edge inscription (magnification x 6) of Simon Cromwell crown (Eric P. Newman) 11. Partial view of edge of 1700 William III crown (Bank of Canada)

10 PLATE IX VOL. XLVI CASPAR, SIMON AND BLONDEAU

11